In this episode, Igor Tomych speaks with Catherine Woneis, VP of Product at Fingerprint, about how fraud detection really works in a digital-first financial system. Catherine draws on two decades of experience in analytics and fraud prevention to explain how device intelligence, browser signals, and AI are used to assess risk without turning security into surveillance.

The conversation moves beyond buzzwords and marketing narratives to focus on practical questions facing fintech teams today, from privacy-preserving authentication to the rise of automated attacks and agentic AI.

Key Insights from the Conversation #

Fraud detection starts before authentication:

Fraud prevention no longer begins when a user enters a username and password. Device and browser signals can reveal anomalies the moment someone lands on a website or opens an app. These early signals help organizations decide whether to allow a smooth experience or introduce additional checks, even if the credentials later appear valid.

Device fingerprinting is about consistency, not tracking:

Device intelligence is often misunderstood as cross-site tracking. In reality, its purpose is to verify whether a visitor’s device behaves consistently over time. When configurations, locations, or technical signals change in ways that do not align, they indicate possible manipulation or fraud, while still respecting privacy boundaries.

AI evaluates intent, not just anomalies:

Machine learning helps distinguish normal variation from suspicious behavior by comparing current signals with known patterns of legitimate and fraudulent activity. Rather than making hard decisions, AI provides risk context, enabling fraud systems to respond proportionally instead of relying on rigid rules.

Automation is no longer inherently malicious:

As agentic AI and automated browsing grow, not all bots are bad. Some act legitimately on behalf of users, while others are designed to exploit systems. The challenge for fraud teams is learning how to differentiate helpful automation from fraudulent automation at scale.

Why Listen #

This episode offers a grounded, technical perspective on fraud prevention from someone building these systems in practice. Listeners will gain a clearer understanding of how device intelligence, AI, and human behavior intersect, and why digital trust depends less on single signals and more on context, consistency, and intent.

Expert Brief #

Catherine Woneis is Vice President of Product at Fingerprint with over 20 years of experience in data analytics and fraud prevention. She has worked across text, visual, and blockchain analytics, focusing on using device intelligence and AI to detect fraud, assess intent, and strengthen digital trust while preserving user privacy.